W3C home > Mailing lists > Public > www-rdf-interest@w3.org > April 2002

Re: Disambiguation; keeping the "U" in "URI"

From: Mark Baker <distobj@acm.org>
Date: Wed, 24 Apr 2002 14:47:18 -0400
To: Miles Sabin <msabin@interx.com>
Cc: www-rdf-interest@w3.org
Message-ID: <20020424144718.E13270@www.markbaker.ca>
On Wed, Apr 24, 2002 at 03:18:45PM +0100, Miles Sabin wrote:
> Mark Baker wrote,
> > Re disambiguation, as I've said repeatedly (don't people believe me?
> > 8-), HTTP has the Content-Location header for doing just this.
> Right, this helps but, IMO, it doesn't solve the whole problem, and
> it's only one mechanism amongst many.
> One of the major problems is that it seems to require an HTTP 
> retrieval, which might be inappropriate, and that the server be able 
> to make the decision about the correct disambiguation, which might
> not always be the case.

That's true, but similar semantics can encapsulated within a
separate property that can be used in other contexts.

<http://www.markbaker.ca/index.html> foo:variantRepresentationOf

Content-Location is a transient assertion, since its subject is a
sequence of bytes found in an HTTP response.  So it wouldn't be useful
useful to capture that exactly in a generic manner (note to self;
update RestRDF).

> This seems particularly unfortunate in the car vs. document case. If
> I want to make an assertion about a car I surely don't want to have
> to attempt a retrieval on it's URI only to be told by the server to
> refer to an associated document. The retrieval is innapropriate in
> the first place, but, in the absence of a retrieval, where's the
> Content-Location: header to do the work of disambiguation?

I agree that the same (or similar) relationship can and should be
communicated in other ways.

> Mark, you've been so insistent that http://www.markbaker.ca/ 
> designates you, that I think it's reasonable to believe that it's
> passed into common use as one of your names, at least around here. 
> Which means that someone might ask me to "Buy http://www.markbaker.ca/ 
> a beer". Context and the nature of the beer-buying operation means
> that I'm unlikely to do anything daft with a document in response to
> such a request, but it's quite clear that neither HTTP nor Content-
> Location: do any work the process.

Well, if I changed my web server to redirect to http://www.markbaker.com
(note the ".com" - i.e. not me), then I would expect that there'd be
confusion about who exactly "http://www.markbaker.ca" defines.  So it is
part of the process, IMO.

Mark Baker, Chief Science Officer, Planetfred, Inc.
Ottawa, Ontario, CANADA.      mbaker@planetfred.com
http://www.markbaker.ca   http://www.planetfred.com
Received on Wednesday, 24 April 2002 14:40:27 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:07:40 UTC